This is the accessible text file for GAO report number GAO-06-824
entitled 'Treasury's Bank Enterprise Award Program: Impact on
Investments in Distressed Communities Is Difficult to Determine, but
Likely Not Significant' which was released on July 31, 2006.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
Report to Congressional Committees:
United States Government Accountability Office:
GAO:
July 2006:
Treasury's Bank Enterprise Award Program:
Impact on Investments in Distressed Communities Is Difficult to
Determine, but Likely Not Significant:
Treasury's Bank Enterprise Award Program:
GAO-06-824:
GAO Highlights:
Highlights of GAO-06-824, a report to congressional committees
Why GAO Did This Study:
Established in 1994, the Department of the Treasury’s Bank Enterprise
Award (BEA) program provides cash awards to banks that increase their
investments in community development financial institutions (CDFI) and
lending in economically distressed communities. CDFIs are specialized
institutions that provide financial services to areas and populations
underserved by conventional lenders and investors. In 2005, Treasury
provided nearly $10 million in BEA awards.
The BEA program has faced longstanding questions about its
effectiveness and experienced significant declines in funding in recent
years. This report (1) examines the extent to which the BEA program may
have provided banks with financial incentives and (2) assesses the BEA
program’s performance measures and internal controls.
To complete this study, GAO reviewed relevant award data; interviewed
Treasury, bank, and CDFI officials; and assessed the BEA program’s
performance measures and internal controls against GAO’s standards for
effective measures and controls.
What GAO Found:
The extent to which the BEA program may provide banks with incentives
to increase their investments in CDFIs and lending in distressed
communities is difficult to determine, but available evidence GAO
reviewed suggests that the program’s impact has likely not been
significant. Award recipients GAO interviewed said that the BEA program
lowers bank costs associated with investing in a CDFI or lending in a
distressed community, allowing for increases in both types of
activities. However, other economic and regulatory incentives also
encourage banks to undertake award-eligible activities, and it is
difficult to isolate and distinguish these incentives from those of a
BEA award. For example, banks may have economic incentives to lend in
distressed communities because of the potential profitability of such
lending. Although it is difficult to determine the BEA program’s
impact, available evidence suggests that the impact likely has not been
significant. For example, the size of a BEA award for large banks
(which was .0004 percent of assets in 2005) suggests that a BEA award
does not have much influence on such banks’ overall investment and
lending decisions (see figure). However, BEA awards may allow large
banks to incrementally increase their award-eligible investments and
lending.
The BEA program’s performance measures likely overstate its impact, and
GAO identified weaknesses in certain program internal controls. To
assess the BEA program’s performance, Treasury, among other measures,
annually aggregates the total reported increase in CDFI investments and
distressed community loans by all applicants but does not account for
other factors, such as economic and regulatory incentives that also
affect bank decisions. GAO also found that Treasury has limited
controls in place to help ensure that BEA program applications contain
accurate information. In particular, Treasury provides limited guidance
to application review staff to identify potential errors and does not
require the reviewers to completely document their work. As a result,
GAO found that the BEA program is vulnerable to making improper
payments.
Table: Average BEA Award as Percentage of Large Banks; Assets [A], 2003
through 2005:
Year: 2003;
Number of banks[B]: 21;
Average award as percentage of total assets: .0005.
Year: 2004;
Number of banks[B]: 17;
Average award as percentage of total assets: .0004.
Year: 2005;
Number of banks[B]: 22;
Average award as percentage of total assets: .0004.
Source: GAO analysis of Treasury data.
[A] Large banks, for purposes here, are those with total assets of $1
billion or more.
[B] Large banks received 43 percent of all BEA award dollars in 2003, 8
percent in 2004, and 38 percent in 2005.
[End of table]
What GAO Recommends:
GAO recommends that Treasury strengthen its internal controls to ensure
proper award payments. Treasury disagreed with aspects of GAO’s
analysis but agreed to implement the recommendation.
[Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-06-824].
To view the full product, including the scope and methodology, click on
the link above. For more information, contact George A. Scott at (202)
512-5932 or scottg@gao.gov.
[End of Section]
Contents:
Letter:
Results in Brief:
Background:
The BEA Program Reportedly Produces Benefits, but Available Evidence
Suggests That the Program's Impact Has Likely Not Been Significant:
The BEA Program's Performance Measures Likely Overstate Its Impact, and
Treasury's Internal Controls to Ensure Proper Award Payments Have
Weaknesses:
Conclusions:
Recommendation for Executive Action:
Agency Comments and Our Evaluation:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: Comments from the Department of the Treasury:
GAO Comments:
Appendix III: GAO Contact and Staff Acknowledgments:
Tables:
Table 1: Percentage of Reported Increase in Award-Eligible Activities,
Fiscal Year 2005 and 2006:
Table 2: Average BEA Award as a Percentage of Large Banks' Assets, 2003
through 2005:
Abbreviations:
BEA: Bank Enterprise Award:
CDFI: community development financial institution:
CRA: Community Reinvestment Act of 1977:
OMB: Office of Management and Budget:
Treasury: U.S. Department of the Treasury:
United States Government Accountability Office:
Washington, DC 20548:
July 31, 2006:
The Honorable Christopher Bond:
Chairman:
The Honorable Patty Murray:
Ranking Minority Member:
Subcommittee on Transportation, Treasury, the Judiciary, Housing and
Urban Development, and Related Agencies:
Committee on Appropriations:
United States Senate:
The Honorable Joe Knollenberg:
Chairman:
The Honorable John W. Olver:
Ranking Minority Member:
Subcommittee on Transportation, Treasury, and Housing and Urban
Development, the Judiciary, the District of Columbia and Independent
Agencies:
Committee on Appropriations:
House of Representatives:
Established in 1994, the Department of the Treasury's (Treasury) Bank
Enterprise Award (BEA) program was designed to provide financial
incentives for FDIC-insured banks and thrifts (hereafter referred to as
banks) to increase their investments in community development financial
institutions (CDFI)[Footnote 1] and lending within eligible distressed
communities as defined by statutory and regulatory
requirements.[Footnote 2] CDFIs are private for-profit or not-for-
profit financial institutions that provide financial services (e.g.,
loans) to communities traditionally underserved by conventional lenders
and investors and that Treasury may certify for participation in the
BEA program and other related programs.[Footnote 3] CDFIs include
community development banks, which may receive BEA awards because they
are FDIC-insured; credit unions, which are ineligible for BEA awards
because they are not FDIC-insured; loan funds; and venture capital
funds.[Footnote 4] Due to statutory and regulatory requirements,
community development banks, which tend to be small institutions,
receive relatively larger BEA awards for increasing certain award-
eligible investments and lending compared to traditional
banks.[Footnote 5] In providing banks with incentives to increase their
award-eligible activities, the BEA program seeks to build the financial
capacity of CDFIs, so they may better serve their customers, and the
availability of direct lending within distressed communities.
However, the BEA program has faced long-standing questions about its
effectiveness and experienced significant funding declines in recent
years. A 1998 GAO report, as well as a 2002 review by the Office of
Management and Budget (OMB), both questioned the extent to which the
BEA program provided banks with financial incentives to increase their
award-eligible activities.[Footnote 6] For example, we and OMB stated
that the Community Reinvestment Act of 1977 (CRA) provides banks with
incentives to make similar investments and loans that the BEA program
awards and that it can be difficult to distinguish CRA's incentives
from those of a BEA award.[Footnote 7] Further, from fiscal years 2000
through 2005, BEA program funding declined from over $46 million to
about $10 million, the number of award recipients declined from 159 to
53, and Treasury has increasingly been unable to award all qualified
applicants.[Footnote 8] The average BEA award amount also dropped from
almost $292,000 to about $187,000 during the period.
Noting concerns about funding reductions to the BEA program and other
related programs within Treasury, as well as the lack of a recent third-
party evaluation, a fiscal year 2006 report by the Senate Committee on
Appropriations requires us to assess the BEA program, particularly the
extent to which it affects bank behavior in providing financial
services to distressed communities.[Footnote 9] As agreed with
committee staff, this review also includes an assessment of certain
aspects of Treasury's administration of the BEA program. Accordingly,
this report (1) examines the extent to which the BEA program may have
provided banks with financial incentives to increase their investments
in CDFIs and lending in distressed communities and (2) assesses the BEA
program's performance measures and certain internal controls designed
to ensure proper award payments.
To conduct our work, we reviewed relevant program statutes,
regulations, guidelines, memorandums, and reports; interviewed Treasury
officials regarding the BEA program's impact and administration;
interviewed CDFI trade associations regarding their views of the
program; and interviewed a nonprobability sample of nine BEA award
recipients and five CDFI beneficiaries participating in the fiscal year
2005 round of awards.[Footnote 10] While results from these interviews
cannot be projected to the entire population of BEA award recipients
and beneficiaries, we selected these recipients and beneficiaries for
interviews to assure variation on a range of characteristics, including
differing asset sizes, frequency of program participation, award-to-
asset percentages, and CDFI type. Our interviews with award recipients
included both community development banks and traditional banks. We
also assessed the BEA program's performance measures and internal
controls against our standards for effective measures and controls.
We conducted our work from October 2005 through July 2006 in
Washington, D.C., in accordance with generally accepted government
auditing standards. Appendix I provides a description of our scope and
methodology in greater detail.
Results in Brief:
The extent to which the BEA program may provide banks with incentives
to increase their investments in CDFIs and lending in distressed
communities is difficult to determine, but available evidence we
reviewed suggests that the program's impact likely has not been
significant. According to Treasury officials and some BEA award
recipients we interviewed, the BEA program produces a range of
benefits, such as lowering bank costs associated with investing in a
CDFI or lending in a distressed community, which encourages and allows
banks to increase both types of activities. According to Treasury
officials, the BEA program has also encouraged partnerships between
banks and CDFIs. However, independently evaluating and isolating the
BEA program's impact is difficult because other economic and regulatory
incentives also affect bank behavior. For example, banks have economic
incentives to lend in distressed communities because BEA-eligible loans
can be profitable. In addition, CRA provides banks with a regulatory
incentive to undertake award-eligible activities. In accordance with
CRA, federal regulators examine and assess banks based on their efforts
to provide financial services (e.g., investments in CDFIs or loans in
distressed communities) in all areas of the community they serve and
may consider inadequate compliance when reviewing a bank's application
to merge or expand operations. Moreover, even when not accounting for
other economic and regulatory incentives, BEA awards for large banks
may be small and, therefore, may not have much influence on their
overall investment and lending decisions, although the awards may
provide such banks with the capacity to incrementally increase their
award-eligible activities. In addition, until 2003, BEA awards may have
provided certain community development banks with incentives to benefit
financially from activities that were inconsistent with program goals,
and available studies indicate that certain CDFIs have been able to
raise an increased amount of capital from banks concurrent with recent
declines in BEA program funding and participation.
The BEA program's performance measures likely overstate the program's
impact; in addition, we identified weaknesses in certain BEA program
internal controls. To assess the BEA program's performance, Treasury,
among other measures, annually aggregates the total reported increase
in CDFI investments and distressed community loans by all applicants
and attributes this increase solely to the BEA program. For example,
Treasury attributed a reported $100 million increase in applicants'
CDFI investments and distressed community loans to the $10 million in
BEA awards it distributed in 2005. Because this and similar BEA program
performance measures do not isolate the prospect of BEA award receipt
from other economic and regulatory incentives, such as loan
profitability and CRA requirements, they likely attribute more
influence to the program than can be substantiated. Furthermore, we
identified weaknesses in the BEA program's system of internal control,
which increase its vulnerability to improper payments. Specifically, we
found that Treasury has limited controls in place to ensure that BEA
applications contain accurate information upon which to make award
determinations (i.e., bank-financed properties are located in eligible
distressed communities as defined by statutory and regulatory
requirements). We also found that Treasury provides limited guidance to
its application review staff to identify potential errors in the
reporting of a financed property's location and does not require the
reviewers to completely document their work.
This report recommends that Treasury revise its guidance to application
review staff and require staff to document their work to help ensure
that errors in the reporting of property location are identified and
the risk of improper payments is minimized. Treasury provided written
comments on a draft of this report that are reprinted in appendix II.
In its comments, Treasury agreed with our conclusion that determining
the BEA program's impact is difficult, but disagreed with certain
aspects of our analysis. For example, Treasury said that our
examination of the BEA program's impact on bank behavior bases many of
its conclusions on information that is overly general, outdated, or
developed for purposes other than to evaluate the BEA program. Treasury
also said that we did not adequately consider evidence the department
provided regarding the BEA program's impact. We believe the information
and evidence used to support our conclusions is appropriate and
continue to conclude that the BEA program's impact on bank behavior has
likely not been significant. Treasury did agree to implement our
report's recommendation. Treasury's comments and our evaluation of them
are discussed in greater detail at the end of this report. Treasury
also provided technical comments that we have incorporated, as
appropriate.
Background:
The BEA program's goals are to encourage banks to increase their
investments in CDFIs and lending and other financial services in
distressed communities.[Footnote 11] Unlike grant programs, which are
usually prospective--meaning they award applicants based on their plans
for the future--the BEA program is retrospective, awarding applicants
for activities they have already completed. Under the program's
authorizing statute, BEA award recipients are not limited in how they
may use their award and, therefore, may use their award proceeds in any
manner they deem fit.
To encourage increased investment and lending, the BEA program awards
applicants on the basis of their increased activities from one year
(known as the baseline year) to the next (the assessment
year).[Footnote 12] For example, for the fiscal year 2005 round of
awards, calendar year 2003 was the baseline year and calendar year 2004
was the assessment year. When applying for awards, applicants may
submit an application for any of the following three award categories:
(1) CDFI-related activities, (2) distressed community financing
activities, and (3) service activities. CDFI-related activities are
primarily investments in CDFIs, such as equity investments (including
grants and equitylike loans), loans, and insured deposits. Distressed
community financing activities are primarily loans, such as affordable
housing loans, small-business loans, commercial real estate loans, and
education loans. Service activities include the provision of financial
services such as check-cashing or money order services, electronic
transfer accounts, and individual development accounts.
Pursuant to statutory and regulatory requirements, BEA awards are
percentage matches of an applicant's reported increase in activities;
that is, banks qualify for a BEA award equal to the sum of the
percentage increase in the three program areas. For equity investments
in CDFIs, the percentage match for both community development banks and
traditional banks is the same--15 percent (see table 1). However,
community development banks are eligible to receive awards three times
higher than traditional banks for increasing CDFI support activities
(e.g., increasing insured deposits in other CDFIs) or increasing their
lending and service delivery in distressed communities. For distressed
community financing activities, a priority factor of 3.0 or 2.0 is
assigned to each type of eligible loan a BEA applicant originates--for
example, a small-business loan is assigned 3.0 and an affordable
housing development loan is assigned 2.0. The change in award-eligible
activity (i.e., the increase in lending from the baseline to the
assessment year) is multiplied by the applicable priority factor, and
the result (or weighted value) is then multiplied by the applicable
award percentage, yielding the award amount for that particular
activity.
Table 1: Percentage of Reported Increase in Award-Eligible Activities,
Fiscal Year 2005 and 2006:
Percent.
BEA-eligible activity: Community development bank;
CDFI-related activities: Equity investments (includes grants and
equitylike loans): 15;
CDFI-related activities: Support activities (includes insured
deposits): 18;
Distressed communities financing activities: 9;
Service activities: 9.
BEA-eligible activity: Traditional bank;
CDFI-related activities: Equity investments (includes grants and
equitylike loans): 15;
CDFI-related activities: Support activities (includes insured
deposits): 6;
Distressed communities financing activities: 3;
Service activities: 3.
Source: GAO.
[End of table]
To illustrate how the BEA program works, suppose a community
development bank that did not have any investments in other CDFIs or
loans in eligible distressed communities during the baseline year.
During the assessment year, the bank makes the following investments or
loans in CDFIs: $300,000 in insured deposits in three community
development credit unions (three insured certificates of deposits of
$100,000 each), $500,000 in small-business loans, and $1 million in
affordable housing development loans in distressed communities (total
increased investments and loans of $1.8 million). Under this example,
the bank would be eligible for a BEA award totaling $369,000[Footnote
13] (a 20.5 percent return on investment).[Footnote 14] Under the same
scenario, a traditional bank would be eligible for a BEA award of
$123,000 (or a return on investment of 6.8 percent).[Footnote 15]
According to Treasury officials, the BEA program is seasonal and
employs the equivalent of about six staff annually, who work on the
program on an as-needed basis. A program manager oversees the BEA
program on a day-to-day basis. During the program's peak application
season, Treasury reassigns roughly 10 staff members from other job
responsibilities to review BEA applications over a period of
approximately 10 business days. During fiscal year 2005, it cost
approximately $1.2 million to administer the BEA program. These costs
are composed of personnel compensation, information technology, and
administrative contracting services, among other costs.
CRA requires federal bank regulators to assess how well the banks they
regulate meet the credit needs of all areas of the community they
serve, including low-and moderate-income areas (insofar as is
consistent with safe and sound operations) and to take this performance
into account when considering a bank's request for regulatory approval
of a regulated action, such as opening a new branch or acquiring or
merging with another bank. Federal regulators conduct examinations for
compliance with CRA requirements on a frequency that varies depending
on an institution's size and prior rating.[Footnote 16] When conducting
examinations, regulators check to see whether a bank's CRA compliance
activities are an ongoing part of the bank's business and generally
apply three tests to make this determination:[Footnote 17]
² A lending test evaluates the number, amount, and income and
geographic distribution of a bank's mortgage, small business, small
farm, and consumer loans.
² An investment test evaluates a bank's community development
investments, including its investments in CDFIs.
² A service test evaluates a bank's retail service delivery operations,
such as branches and low-cost checking services.
Upon completing examinations, regulators assign one of four ratings to
a bank: outstanding, satisfactory, needs improvement, or substantial
noncompliance.
The BEA Program Reportedly Produces Benefits, but Available Evidence
Suggests That the Program's Impact Has Likely Not Been Significant:
Treasury officials and some BEA award recipients we interviewed said
that the BEA program provides banks with incentives to increase their
investments in CDFIs and lending in distressed communities. However,
determining the program's impact is difficult because other economic
and regulatory incentives also encourage banks to undertake award-
eligible activities. Although it is difficult to determine the BEA
program's impact, the available evidence we reviewed suggests that the
program's impact has likely not been significant. For example, for
large banks, a BEA award (when compared with total bank assets) is
small and likely not large enough to have much influence on such banks'
overall investment and lending decisions. Other evidence also indicates
that the BEA program's impact has likely not been significant. In
particular, until 2003, BEA awards may have provided certain community
development banks with incentives to benefit financially from
activities that were inconsistent with BEA program goals, and available
studies indicate that certain CDFIs have been able to raise an
increased amount of capital from banks, while BEA program funding and
participation have declined.
According to Treasury Officials and Some Award Recipients, the BEA
Program Produces a Range of Benefits:
According to Treasury officials and some award recipients, the BEA
program allows award recipients to increase their lending and
investment levels beyond those that would occur without the program.
Award recipients we interviewed stated that one of the program's main
benefits is reduced transaction costs. Transaction costs are primarily
the time and expense associated with researching markets or borrower
qualifications and underwriting loans within distressed communities.
Award recipients stated that transaction costs are higher in distressed
communities than in other communities because, for example, loans are
typically smaller (thus generating less interest income) and have a
higher risk of default. Because BEA awards are in cash, award
recipients said that award proceeds can be used to provide more loans,
on more favorable terms, than are otherwise possible. Award recipients
said that such an arrangement benefits both BEA award recipients and
loan borrowers.
Another benefit that award recipients cited is the formation of
partnerships between banks and other financial institutions, including
CDFIs. When investing in a CDFI--the activity awarded with the highest
payout--applicants identify and select a CDFI in which to invest, such
as a community development bank, credit union, loan fund, or venture
capital fund. According to officials from banks and CDFIs, the
resulting investment in the CDFI produces two benefits. First, the
investment increases the CDFI's capacity by providing it with capital,
often at below-market rates, which in turn allows the CDFI to provide
more loans in distressed communities. Second, according to one CDFI
official we interviewed, the partnership allows traditional banks to
learn about and understand the work of CDFIs. For example, the CDFI
official we interviewed noted that the partnership formed through the
BEA program allowed officials from a traditional bank to sit on the
CDFI's board of directors, which exposed the traditional bank officials
to the products and services of the CDFI. When initially established,
Treasury intended the BEA program to encourage traditional banks to
become involved in community development banking activities by, for
example, investing in a CDFI or lending in a distressed community.
A third benefit of the BEA program, according some award recipients we
interviewed, is the provision of capital needed to help the community
development banking industry grow and develop during its early years
and sustain its level of operations today. An official representing the
community development banking industry noted that there were only three
Treasury-certified community development banks in the mid-1990s when
the BEA program began, but today there are over 50 such banks, growth
the official attributes to the BEA program. Some award recipients we
interviewed also stated that award proceeds have allowed them to
sustain their current level of operations within distressed
communities, where, as previously noted, transaction costs are higher
than in other areas. Accordingly, the BEA program is said to help
community development banks remain true to their core missions of
serving the financing and developmental needs of their community.
Isolating the BEA Program's Impact from Other Existing Economic and
Regulatory Incentives Remains Difficult:
Independently evaluating and isolating the BEA program's impact on bank
investment and lending decisions is difficult because other economic
and regulatory incentives also affect bank behavior. In 1998, we
reported that the prospect of receiving a BEA award, while one factor,
was not always the primary reason banks undertook award-eligible
activities.[Footnote 18] In 2000, the Federal Reserve Board completed a
survey providing additional evidence that loan profitability can be an
important factor in banks' community development lending
decisions.[Footnote 19] This survey, which focused on the performance
and profitability of CRA-related lending, found that a majority of
respondents' community development loans were profitable. The survey
also found that a majority of respondent's CRA special lending
programs, which target low-income borrowers and areas, were
profitable.[Footnote 20] Because community development loans can be
profitable, as noted in the Federal Reserve Board's survey, banks have
economic incentives to make these loans even without the incentive of
potentially receiving a BEA award.
In addition to economic incentives, regulatory incentives can also
encourage banks to undertake award-eligible activities. In our 1998
report, we found that compliance with CRA was a major reason banks made
investments in CDFIs and loans in distressed communities. CRA
incentives may be particularly strong for banks that plan to open a new
branch or merge with other banks because federal regulators may
consider inadequate compliance when reviewing banks' requests to merge
with other banks or expand their operations. However, Treasury
officials said that the BEA program provides banks with more targeted
incentives than CRA requirements do. For example, the officials said
that the BEA program provides banks with incentives to provide
financial services in the most distressed communities--communities that
banks are not required to service in their efforts to comply with CRA.
To obtain feedback on the BEA program's design and implementation,
Treasury has conducted surveys of BEA program applicants. Treasury's
most recent survey, conducted in 2002, suggests that both the BEA
program and CRA requirements are responsible for banks' increased
investments in CDFIs and lending in distressed communities. For
example, the 2002 survey of 115 program applicants found that both the
prospect of a BEA award and credit for CRA compliance motivated banks
to undertake many CDFI-related activities, including providing CDFIs
with loans, grants, and technical assistance, but found that the BEA
program contributed toward the development of new financial products.
The survey also found that, in many cases, neither the BEA program nor
credit for CRA compliance motivated banks to lend in distressed
communities. Rather, the banks reported making loans in distressed
communities because such lending is part of their community development
mission or part of their everyday business activities.
Available Evidence Suggests That the BEA Program's Impact Has Likely
Not Been Significant:
Although it is difficult to determine the BEA program's impact, the
available evidence we reviewed suggests that the program's impact has
likely not been significant for large traditional banks, although it
may allow for incremental increases in award-eligible activities. The
available evidence also suggests that the BEA program may have provided
some community development banks with incentives to benefit financially
without furthering program goals. Further, available studies we
reviewed indicate that some CDFIs have raised an increased amount of
capital from banks while BEA program funding and participation have
declined. Specifically, we found the following:
² For large traditional banks, as noted in our 1998 report, BEA awards
are likely not large enough to provide a meaningful financial
incentive. As shown in table 2, the size of a BEA award when compared
with the assets of large traditional banks (those with over $1 billion
in assets) was .0004 percent of assets in 2005. For these banks, the
prospect of receiving a BEA award, independent of any economic and
regulatory incentives the banks may have, is unlikely to serve as a
significant financial incentive for increased CDFI investment or
distressed community lending. However, BEA awards may provide large
traditional banks with the capacity to incrementally increase their
award-eligible activities, offset some of the cost associated with
doing so, and increase the profits of related lines of business. Large
traditional banks may also derive public and community relations value
from receiving a BEA award that outweighs its financial benefit.
Table 2: Average BEA Award as a Percentage of Large Banks' Assets,A
2003 through 2005:
Year: 2003;
Number of banks[B]: 21;
Average award as percentage of total assets: .0005.
Year: 2004;
Number of banks[B]: 17;
Average award as percentage of total assets: .0004.
Year: 2005;
Number of banks[B]: 22;
Average award as percentage of total assets: .0004.
Source: GAO analysis of Treasury data.
[A] Large banks, for purposes here, are those with total assets of $1
billion or more.
[B] Large banks received 43 percent of all BEA award dollars in 2003, 8
percent in 2004, and 38 percent in 2005.
[End of table]
² Until 2003, many BEA program participants engaged in a now-prohibited
practice called deposit swapping that improved their financial
condition without necessarily furthering program goals. According to a
Treasury official, beginning around 1998, a group of about 30 community
development banks began to purchase insured certificates of deposit in
one another--that is, swap deposits--to increase their CDFI investments
and thereby receive BEA awards. At the time, Treasury provided a 33
percent award match for community development banks that increased
their deposits in other community development banks. Following the 2003
prohibition, the percentage of total BEA dollars awarded for CDFI
investments fell substantially--from 87 percent of all BEA dollars
awarded in 2002 to only 18 percent in 2003 (by contrast, total BEA
dollars awarded for increased lending and services in distressed
communities increased from 13 percent in 2002 to 82 percent in 2003).
According to a Treasury official, the prohibition on deposit swapping
was, in fact, the primary reason for the substantial decline in CDFI
investments. This decline suggests that, until 2003, banks may have
been responding to financial incentives that were inconsistent with the
BEA program's goals, which include increasing lending within distressed
communities.
² Community development loan funds have raised an increased amount of
capital from banks, thrifts, and credit unions, while BEA program
funding and bank participation in the program have declined. According
to data from a consortium of CDFIs, community development loan funds--
the most numerous type of CDFI and thus the largest group of potential
BEA program beneficiaries--have continued raising capital from banks,
thrifts, and credit unions concurrent with a decline in funding and
bank participation in the BEA program.[Footnote 21] According to the
consortium's data, the percentage of capital loan funds raised from
banks, thrifts, and credit unions increased from 47 percent in fiscal
year 2003 to 56 percent in fiscal year 2004. As discussed previously,
BEA program funding also declined substantially in recent years from
over $46 million in fiscal year 2000 to about $10 million in fiscal
year 2005. We note that one limitation of the consortium's data for
purposes of this analysis is that it includes credit unions, which are
ineligible for BEA awards. However, an official involved with
completing the studies said that loan funds raised most of the capital
from banks and thrifts, which are eligible for BEA awards. According to
the CDFI consortium, financial institutions are a growing source of
capital for loan funds because loan funds provide a safe investment,
allow banks to earn CRA credit, and are flexible partners.
The BEA Program's Performance Measures Likely Overstate Its Impact, and
Treasury's Internal Controls to Ensure Proper Award Payments Have
Weaknesses:
Treasury's performance measures for the BEA program likely overstate
its impact on bank investments in CDFIs and lending in distressed
communities. In addition, we identified weaknesses in Treasury's system
of internal control for ensuring proper award payments. Specifically,
we found that Treasury has limited controls in place to help ensure
that bank applicants finance properties located in eligible distressed
communities. We found that Treasury also provides limited guidance to
its application review staff to identify potential errors in the
reporting of a financed property's location and does not require the
reviewers to completely document their work.
BEA Program Performance Measures Likely Overstate Program Impact:
To assess the BEA program's performance, Treasury publicly reports bank
applicants' total reported increase in CDFI investments and distressed
community lending.[Footnote 22] To establish targets for this measure,
Treasury assumes a complete, causal linkage between the BEA program and
applicants' increases in award-eligible activities. For example, in
2005, Treasury attributed a reported $100 million increase in award-
eligible activities to BEA awards of approximately $10 million
distributed that year. In reporting results for this measure, Treasury
does not account for other factors that also affect bank lending and
investment decisions, such as loan profitability and CRA compliance. By
not accounting for such factors, Treasury's performance measure likely
overstates the BEA program's impact. As a result, Treasury lacks
accurate information needed to assess program accomplishments and make
changes to ensure that the BEA program is meeting its goals. GAO's
standards for effective performance measures state that measures should
be objective--that is, they should be reasonably free of any
significant bias or manipulation that would distort an accurate
assessment of performance.[Footnote 23]
Treasury internally tracks other BEA program data, but these data also
likely overstate the program's impact. For example, as part of a BEA
application, Treasury requests that applicants provide such data as the
number of full-time equivalent jobs created or maintained and the
number of housing units developed or rehabilitated in distressed
communities. Treasury uses this information to monitor and measure the
BEA program's impact. Similar to its externally reported measure,
Treasury assumes a direct one-to-one correlation between these outcomes
(new jobs and housing units) and the BEA program. Treasury does not
account for external factors, such as economic and regulatory
incentives that could also contribute to an increase in jobs created or
housing units developed. Further, these data are self-reported and,
according to Treasury, not verified. Therefore, they could be subject
to the type of bias and manipulation that would distort an accurate
assessment of performance.
We acknowledge that developing performance measures for the BEA program
is challenging. As stated in our 1998 report, to an extent that neither
we nor Treasury can quantify, banks are receiving awards for
investments and loans they would have made without the prospect of
receiving a BEA award. The available evidence discussed in this report
(e.g., the relatively small size of BEA awards for large banks) further
supports this analysis. While it may have been advisable for Treasury
to attribute less influence to the BEA program when developing its
performance measures, it is not clear that a reliable and appropriate
methodology exists to accurately measure the BEA program's impact on
bank behavior.
Treasury Has Not Established Effective Controls to Help Ensure That
Bank-Financed Properties Are Located in Eligible Distressed
Communities:
According to a Treasury official, one of the most significant risks the
BEA program faces is that applicants may provide inaccurate information
regarding the location of properties financed by their activities. That
is, the potential exists for banks to receive BEA awards based on loans
that finance properties, such as commercial or affordable housing
development loans, that were not located in eligible distressed
communities. While Treasury has established controls to mitigate this
risk, these controls are not fully consistent with federal internal
control standards, which state that policies and procedures, including
appropriate documentation, should be designed to help ensure that
management's directives, such as verification procedures, are carried
out and that appropriate supervisory oversight of established processes
is exercised. Without sufficient controls to help ensure that
properties are located in eligible distressed communities, the BEA
program is vulnerable to making improper payments.
According to a Treasury official, application review staff are to
perform the following procedures to ensure that properties are located
in eligible distressed communities:
² Use an online Treasury system, for all loans of $500,000 or more, to
verify that borrower addressers or, in some cases, properties secured
by the loans (collateral) are located in eligible census tracts
(generally referred to as loan geocoding).
² Geocode a sample of loans valued at $250,000 to $500,000 to verify
that borrower or collateral addresses are located in eligible census
tracts.
Treasury officials said that BEA program application review staff have
identified properties that were not located in eligible distressed
communities. For example, a Treasury official said that, in one case,
the address of the borrower (a developer), which was located in an
eligible distressed community, was given as a basis for the bank to
receive a BEA award.[Footnote 24] However, the official said that the
address of the property under development was not in an eligible
distressed community. The official said that she was familiar with the
area where the property was located and knew that it did not meet
eligibility requirements, which prompted her to do follow-up analysis.
According to the official, Treasury staff disallowed this particular
loan as a basis for the bank to receive a BEA award.
While a Treasury official said that the department has established
controls to mitigate errors in the reporting of property locations, we
identified limitations with the guidance that Treasury provides to its
application review staff. For example, Treasury's guidance states that
for loans of $500,000 or above and for a sample of loans from $250,000
to $500,000, staff should geocode the borrower's address. However, for
development loans where the address of the borrower (such as a
developer) may differ from the address of the property under
development, the guidance does not specifically require staff to
geocode the property address. A Treasury official confirmed that the
department has not provided specific guidance to reviewers on geocoding
property addresses in such instances. As noted previously, Treasury
staff have identified at least one example in which the location of the
borrower was in a distressed community but the location of the property
was not, although this identification was largely because of the
reviewer's familiarity with the area where the property was located. By
not specifying in the guidance that reviewers should geocode property
addresses where appropriate, the potential exists that banks will
receive BEA awards based on erroneous information.
We reviewed two banks' BEA applications for the fiscal year 2004 and
2005 rounds of BEA awards (a total of four applications) to conduct a
limited test of Treasury's implementation of procedures for verifying
certain application data. Each bank in our review received the maximum
$500,000 award in the 2005 funding round. The files we reviewed did not
contain any documentation of the staff's geocoding of property location
data (for loans exceeding $250,000 or $500,000). A Treasury official we
interviewed agreed that the files did not contain any documentation of
the staffs' geocoding effort. Further, our review of Treasury's BEA
application guidance found that the guidance does not establish
specific documentation requirements for the program staff's geocoding
efforts. Without such guidance and documentation requirements, Treasury
management and supervisors, as well as outside reviewers, cannot be
assured that the geocoding is being conducted or that errors in the
reporting of property location are detected.
To assess the potential for improper BEA award payments, we used
Treasury's online geocoding system to determine the locations of
properties contained in the 2004 and 2005 applications for the two
banks. We identified 1 commercial and 5 affordable housing development
loans among these applications, out of a total of 18 such loans with a
value of $250,000 or more, where we had questions as to whether
properties financed by the loans were located in eligible distressed
communities. For example, we identified an affordable housing
development loan of approximately $423,500 that was made to purchase an
apartment building. Our geocoding analysis determined that the address
of the property was not in an eligible distressed community, whereas
the address of the borrower was in a distressed community that could
qualify under certain circumstances. In this case, according to a
Treasury official, the reviewer probably geocoded the address of the
borrower rather than the address of the property. The Treasury official
also suggested that the address of the property may have been in an
eligible distressed community at the time the application was made in
2004. However, our analysis of census data indicates that the relevant
census tract was not an eligible distressed community in 2004.
Consequently, Treasury's decision to provide a BEA award to this bank
may have been based in part on erroneous information.
Conclusions:
Because of other economic and regulatory incentives that also affect
bank behavior, it remains difficult to isolate and determine the BEA
program's impact on banks' decisions to invest in CDFIs and lend in
distressed communities. Treasury's BEA program performance measures do
not provide additional insights into the program's impact because they
assume that all reported increases in eligible investment and lending
occur solely because of the program's financial incentives. However,
based on available evidence we reviewed, it is reasonable to conclude
that the program likely does not provide significant financial
incentives for large banks, due to the typical award's relatively small
size for such institutions. To an extent that is unquantifiable, a
significant percentage of reported large bank increases in CDFI
investments and distressed community loans each year would likely have
occurred without the BEA program. Further, the program also appears to
have provided certain community development banks with financial
incentives and opportunities to benefit financially without furthering
program goals. On the other hand, the BEA program may provide some
banks, including large banks, with additional incentives and capacity
to incrementally increase their award-eligible activities, offer public
and community relations benefits to some award recipients, contribute
to the development of new financial products, and help establish
partnerships between banks and other CDFIs.
Treasury's internal controls to ensure proper award payments are
insufficient. Treasury's guidance to its BEA application review staff
does not require them to geocode property addresses, even though
evidence exists that applications may contain errors in reported
information. The guidance also does not establish standards for
documenting verification efforts. Consequently, the BEA program is
vulnerable to making improper payments.
Recommendation for Executive Action:
To help ensure the integrity of the BEA award payment process, we
recommend that the Secretary of the Treasury revise the guidance for
reviewing program applications so that program staff are required to
(1) geocode property addresses where appropriate and (2) document their
efforts to verify property addresses.
Agency Comments and Our Evaluation:
We provided a draft of this report to the Department of the Treasury
for its review and comment. Treasury provided written comments that are
reprinted (with annotations) in appendix II. In its comments, Treasury
agreed with our conclusion that determining the extent to which the BEA
program provides banks with incentives to increase their investments in
CDFIs and lending in distressed communities remains difficult given the
number of external factors that drive such decisions. However, Treasury
stated that our report bases many of its conclusions on information
that is overly general, outdated, or developed for other purposes and,
as a result, does not reflect an accurate portrayal of the BEA program
or its importance within the banking industry. Treasury also said that
we did not adequately consider evidence the department provided
regarding the BEA program's impact. Treasury did agree to implement our
recommendation that application review staff (1) geocode property
addresses, where appropriate; and (2) document their efforts to verify
property addresses. Further, Treasury stated that it will adopt a
policy requiring applicants to report addresses for transactions;
provide program staff with updated instructions to geocode all
transactions over $250,000 (not just transactions over $500,000, as is
the current practice); and initiate and implement steps to analyze a
statistically significant sample of transactions less than $250,000.
In its comments, Treasury stated that the focus of our report was
inherently flawed. Treasury said our report did not assess, as it
expected, whether the BEA program, as currently structured, is
effective at motivating banks to undertake community development
financing activities they would not normally undertake or, if the
program were found to be ineffective, recommend changes to its
structure. In fact, we did seek to assess whether the BEA program, as
currently structured, is effective at motivating banks to undertake
activities they would not normally undertake. However, as was the case
when we initially evaluated the BEA program in 1998 and as we state in
this report, because of other economic and regulatory incentives that
affect bank behavior, it is difficult to isolate the BEA program's
impact from these other incentives. We note an absence of change in the
banking industry since 1998 that would facilitate isolating the BEA
program's impact for this review. On the contrary, isolating the BEA
program's impact may be more difficult today than in 1998 because the
average BEA award amount and number of banks participating in the
program have declined significantly in recent years. Although isolating
the impact of the BEA program is difficult, we believe available
evidence suggests that its impact has likely not been significant.
Treasury also stated that our report relied on inappropriate
information and data to form conclusions and that we did not consider
other evidence. For example, Treasury stated that none of the studies
cited in the report--including our 1998 report, a 2000 Federal Reserve
survey on CRA-related lending, and two studies by a consortium of
CDFIs--is an explicit evaluation of the BEA program. Treasury also
stated that we undertook only a limited review of current program
participants. Contrary to Treasury's assertions, our 1998 report
includes an assessment of the BEA program. Moreover, the Federal
Reserve survey and reports by a consortium of CDFIs address issues that
we believe are critical to independently evaluating the BEA program's
effectiveness. In particular, the Federal Reserve survey indicates that
community development lending can be profitable, which suggests that a
variety of factors--including economic and regulatory factors--
influence bank lending decisions. The variety of factors that can
influence bank lending decisions increase the difficulties associated
with isolating and determining the BEA program's impact. As discussed
in this report, the data from the consortium of CDFIs also provide
evidence that community development loan funds have been able to raise
an increased amount of capital from banks despite recent declines in
BEA program funding and participation. Regarding our interviews with
program participants, as we note in appendix I, we chose program
participants for interviews based on a variety of characteristics--
including differing bank asset sizes, frequency of program
participation, status as a traditional bank or community development
bank, and CDFI type--to elicit a wide range of views and perspectives
on the BEA program.
Further, Treasury stated that we did not adequately refer to its 2002
survey of BEA program participants in our draft report. Treasury stated
that evidence from the survey clearly demonstrates that the BEA program
plays a role in program applicant investment decisions. While we
recognize that surveys of program beneficiaries can play an important
role in program evaluations, we believe that their results must be
interpreted with caution. For example, survey respondents who are
program beneficiaries have a financial incentive to overstate a
program's impact. To compensate for this limitation, we sought to
obtain and analyze independent evidence, including available studies,
to assess the BEA program's impact. Even so, the findings of Treasury's
2002 survey are consistent with the findings of our report. For
example, our report states that prior to 2003, when deposit swapping
was prohibited, the BEA program may have provided certain community
development banks with incentives to make investments that benefited
them financially but were inconsistent with program goals. In
Treasury's 2002 survey, CDFI deposits was the only category in which a
majority of bank respondents (52 percent) said that the BEA program was
the primary reason they made an award-eligible investment. Overall,
Treasury's 2002 survey indicates that various factors, which include,
but are not limited to, the prospect of receiving a BEA award, motivate
banks' decisions to invest in CDFIs and lend in distressed communities.
In fact, Treasury's 2002 survey found that in many cases, neither the
BEA program nor credit for CRA compliance motivated banks' decisions to
lend in distressed communities. Rather, as we state in our report, the
survey found that respondents undertook lending activities because they
were part of their community development mission or part of their
everyday business activities.
Additionally, Treasury said that some conclusions in the report appear
to reflect a lack of understanding of the BEA program and the banking
industry. Specifically, Treasury stated the following:
² GAO's analysis of the size of a BEA award relative to large banks'
total assets was overly general and did not consider that many banks
(in particular large banks) carry out CDFI financing within specific
lines of business, such as community development business lines. Rather
than comparing a large bank's BEA award amount with its total assets,
as we did, Treasury said a more appropriate and meaningful analysis
would have been to consider the bank's BEA award to the assets of a
particular business line or its relative importance in lowering the
bank's transaction costs. In response to this comment, we added
language to the report that, for large traditional banks, BEA awards
may provide additional capacity to incrementally increase award-
eligible investments and lending, offset some of the costs associated
with doing so, and increase the profits of related lines of business.
In interviews for this report, officials from one large bank said BEA
awards have allowed their bank to provide more loans than they would
have in the program's absence, and officials from another large bank
said BEA awards have allowed their bank to provide loans on more
favorable terms. However, the officials said that other factors, such
as CRA compliance and loan profitability, also influence their
community development lending decisions. Further, officials from both
banks said their banks would continue community development lending in
the BEA program's absence, although officials from one bank said their
bank would continue such lending to a lesser extent. Therefore, we
continue to believe that the BEA program likely does not have a
significant impact on large banks' overall investment and lending
decisions, although there may be an incremental impact.
² GAO's discussion of the now-prohibited practice of deposit swapping
was based on outdated information, as Treasury moved to prohibit this
practice four years ago. Treasury said it did not understand why we
chose to include a discussion of deposit swapping in a report on the
BEA program's current status. In response to this comment, we assert
that our report sought to assess the BEA program's impact on bank
behavior over time, rather than at a single point in time. Thus, we
believe that our discussion of deposit swapping, which focuses on bank
behavior in response to incentives that the BEA program provided until
2003, is appropriate. We note that deposit swapping provides evidence
that, until 2003, the BEA program's impact in encouraging some banks to
make productive investments and loans in distressed communities likely
was not significant. We also note that funding for the BEA program, and
bank participation in it, were highest prior to 2003 when Treasury
prohibited deposit swapping, adding significance to the issue of
deposit swapping and its connection to bank behavior.
² GAO's report failed to mention other important program benefits. In
support of this statement, Treasury cites its 2002 survey in which 19
percent of respondents indicated that the prospect of receiving a BEA
award prompted them to launch innovate financial products, services, or
educational programs to meet the needs of underserved households or
communities. In response to this comment, we revised our report to
reflect this survey finding. Treasury also stated that it would have
been useful if our report studied the underlying data from the
consortium of CDFIs to, among other things, determine the BEA program's
impact in initiating productive relationships between banks and CDFIs.
Our draft report stated that a benefit of the BEA program is that it
encourages partnerships between banks and CDFIs. However, it was not
possible to determine from the CDFI consortium data we reviewed whether
the loan funds cited in the reports formed partnerships with banks
participating in the BEA program. For example, the consortium reports
did not specifically identify the loan funds and banks that were
surveyed for inclusion in the reports. Therefore, based on information
in the reports, we were unable to conduct the types of analyses
Treasury proposes in its comments.
We are sending copies of this report to the Secretary of the Treasury
and other interested congressional committees. We will also make copies
available to others upon request. In addition, the report will be
available at no charge on GAO's Web site at [Hyperlink,
http://www.gao.gov].
If you or your staffs have any questions regarding this report, please
contact me at (202) 512-7215 or scottg@gao.gov. Contact points for our
Offices of Congressional Relations and Public Affairs may be found on
the last page of this report. GAO staff who made key contributions to
this report are listed in appendix III.
Signed by:
George A. Scott:
Acting Director, Financial Markets and Community Investment:
[End of section]
Appendix I: Objectives, Scope, and Methodology:
The objectives of this report were to (1) examine the extent to which
the Bank Enterprise Award (BEA) program may have provided banks with
financial incentives to increase their investments in community
development financial institutions (CDFIs) and lending in distressed
communities and (2) assess the BEA program's performance measures and
certain internal controls designed to ensure proper award payments.
To address our first objective, we reviewed relevant documents and
data, including BEA program statutes, regulations, memorandum,
guidelines, and reports; GAO's 1998 report on the CDFI Fund and BEA
program; a 2000 Federal Reserve Board study on the performance and
profitability of Community Reinvestment Act-related lending,[Footnote
25] and two studies by the CDFI Data Project, which is an industry
consortium that gathers and reports financial data on the CDFI
industry.[Footnote 26] We also interviewed three trade associations
representing various segments of the CDFI industry to obtain their
views on the BEA program. Further, we interviewed a nonprobability
sample of nine BEA award recipients and five CDFI beneficiaries from
the fiscal year 2005 round of BEA awards. We selected these award
recipients and CDFI beneficiaries for interviews based on a range of
characteristics, including differing bank asset sizes, frequency of
program participation, status as a traditional bank or certified
community development bank, and CDFI type. Our sample selection
criteria was intended to obtain a diverse pool of respondents
possessing a range of views and perspectives on the BEA program.
To address our second objective, we interviewed Treasury officials to
obtain information on the BEA program's measures and internal controls.
We compared the program's performance measures to GAO's standards for
effective measures, as outlined in publications we have issued in
connection with the Government Performance and Results Act. We also
compared the BEA program's internal controls to GAO's Standards for
Internal Control in the Federal Government.[Footnote 27] To further
assess the program's internal controls, we reviewed application
documents for two banks that each received multiple BEA awards from
2000 through 2005 and used Treasury's online geocoding system to
determine the locations of properties contained in the 2004 and 2005
applications for the two banks. We also reviewed BEA program
application review guidance.
[End of section]
We conducted our work from October 2005 through July 2006 in
Washington, D.C., in accordance with generally accepted government
auditing standards.
[End of section]
Appendix II: Comments from the Department of the Treasury:
Note: GAO comments supplementing those in the report text appear at the
end of this appendix.
Department Of The Treasury:
Community Development Financial Institutions Fund:
801 Thirteenth Street, NW, Suite 200 South:
Washington, DC 20005:
July 21, 2006:
George A. Scott:
Acting Director, Financial Markets and Community Investments:
Government Accountability Office:
441 G St., N W
Washington, DC 20548:
Dear Mr. Scott:
The Community Development Financial Institutions Fund (CDFI Fund) has
reviewed the Government Accountability Office (GAO) report on the Bank
Enterprise Award (BEA) Program ("Treasury's Bank Enterprise Award
Program: Impact on Investments in Distressed Communities is Difficult
to Determine, But Likely Not Significant for Some Banks"). The CDFI
Fund agrees with the GAO's conclusion that measuring the extent to
which the BEA program provides an incentive to banks to increase their
investments in CDFIs and lending in distressed communities is
difficult, given the number of external factors that drive these
decisions.
The CDFI Fund also agrees with GAO's observations that the BEA Program
has provided a critical source of capital to many smaller CDFI banks,
which is certainly consistent with the statutory intent of the BEA
Program. However, as discussed in greater detail below, the CDFI Fund
believes that the GAO bases many of its conclusions upon information
that is overly general, outdated, or developed for other purposes than
to evaluate the BEA Program. As a result of these flawed methodologies,
we believe the report does not reflect an accurate portrayal of the BEA
Program and/or its importance in connection with the community
development activities of the banking industry.
CDFI Fund Comments:
1. The focus of the report was inherently flawed.
The CDFI Fund had expected the GAO report to opine as to whether the
program as currently structured was effective in motivating banks to
undertake community development financing activities that they would
not undertake in the normal course of business and, if it was not, to
make recommendations on how to change the program structure to increase
its effectiveness. Instead, the GAO report, in our judgment, critiques
the difficulty of measuring the impact of the BEA program on bank
behavior, yet still makes conclusions suggesting the program is having
minimum impact. While we agree that there are inherent limitations to
analyzing bank behavior, we are also well aware of available evidence
indicating the positive impact that the BEA Program has had on the
community development activities of many banks.
We shared evidence of these positive impacts with GAO during interviews
and through written materials; yet, we see little reference to this in
the report. For example, we provided the GAO with the results of a
comprehensive survey of program participants undertaken by the CDFI
Fund in FY 2002. The results of this survey, clearly demonstrate that
the prospect of receiving a BEA Program award plays a role in an
applicant's decision to engage in BEA-eligible investments in CDFIs,
even when the positive responses regarding the now-prohibited practice
of making deposits in CDFIs is disregarded. Not including the responses
relating to deposits, for each other eligible CDFI-Related Activity,
between 43% and 100% of respondents indicated this was the case, and
these facts seem to be given little weight or credence by the GAO.
2. The information and data relied upon by GAO to form its conclusions
was inappropriate, and other evidence was not considered.
In undertaking its evaluation, the GAO focused on information that was
overly general (e.g., the size of the award relative to bank asset
size), outdated (the now-prohibited practice of "deposit swapping"), or
developed for other purposes (e.g., a study of the profitability of
community development lending). In several parts of the report, the GAO
refers to "the available evidence" ("The BEA Program reportedly
produces benefits but available evidence suggests."). As listed in
Appendix I, the evaluative evidence seems to consist of only two
studies, both significantly outdated.
One is the GAO's 1998 report on the monitoring and measurement
approaches the CDFI Fund used in connection with each of its programs,
among which is BEA, and a 2000 Federal Reserve Board study on the
performance and profitability of CRA-related lending. In addition two
CDFI Data Project reports are referenced. None of these sources is a
study or explicit evaluation of the BEA Program. Moreover, the GAO
undertook only a limited survey of current program participants. The
CDFI Fund had hoped the GAO would conduct a larger scale review of
participant and/or would build from the CDFI Fund's 2002 survey of
program participants. Notwithstanding the small sample size the GAO
chose, the results of interviews that the GAO conducted with nine BEA
Program awardees appear to support the CDFI Fund's previous findings
that the BEA Program has positive impact on the community development
activities of many banks. However, it appears that the GAO chose to not
consider this evidence in their report.
3. Some conclusions appear to reflect a lack of understanding of the
BEA Program and the banking industry.
The CDFI Fund believes that the GAO's assertion that the BEA Program
has not been effective for some banks, because the size of a BEA
Program award is small relative to the asset size of large traditional
banks, misses the point of the program and reflects a lack of
understanding of the banking industry. For a traditional bank with
assets in the billions of dollars, comparing the size of the BEA
Program award (which averages about $200,000) to its total assets is
not a suitable measure.
In actual practice, large traditional banks use BEA Program awards to
mitigate risk and improve the economies of individual CDFI-related
transactions. Many banks (and, in particular, large banks) carry out
CDFI financing within a specific line of business (community
development activities). These specific business units must demonstrate
profitability. Therefore, a more appropriate and meaningful analysis
would have been to consider the size of a BEA Program award relative to
the assets of this particular line of business within a bank, or its
relative importance in lowering the transaction costs.
Another indicator used by the GAO to assess program effectiveness was
the now-prohibited practice of "deposit swapping" among CDFI banks. As
the CDFI Fund pro-actively moved to prohibit this practice four years
ago (in the regulatory revisions effective with the FY 2003 funding
round), it is unclear to us why the GAO has chosen to mention the issue
in a report on the current status of the BEA Program. As we explained
to the GAO, the practice was engaged in by a limited subset of
applicants and was prohibited for the exact reason the GAO cites: it
was inconsistent with the goals of the program.
The GAO report also notes that it is difficult to isolate and
distinguish other economic and regulatory incentives from those of the
BEA Program award. The CDFI Fund acknowledges that bank behavior is
affected by such influences - especially that of the CRA - but we
believe that it is possible to make distinctions between the BEA
Program and the CRA. Congress established the BEA Program as a source
of direct monetary assistance to banks for activities that are
complementary to their regulatory CRA goals.
The BEA Program, however, was not intended as a mechanism for a bank to
improve its CRA rating. While BEA-eligible transactions typically
receive favorable consideration during a bank's CRA examination, banks
do not need to participate in BEA eligible transactions in order to
comply with the CRA. Indeed, while participation in the BEA Program can
be an important part of a successful CRA strategy, it is too highly
targeted - and the size of the awards too small - to be the primary
focus of any bank's CRA strategy.
The GAO failed to consider an important distinction between the BEA
Program and the CRA: the different levels of geographic targeting. In
evaluating banks' CRA performance, the regulatory agencies take into
consideration activities in Low and Moderate Income census tracts
within their respective service areas. Using the 2000 census, there are
18,379 such CRA census tracts throughout the country. The BEA Program,
on the other hand, employs a stricter standard. Specifically, a BEA-
qualified Distressed Community is a census tract or a group of census
tracts that have at least a 30 percent Poverty Rate and an Unemployment
Rate of at least 1.5 times the national average. Using 2000 Census
data, there are 2651 census tracts that qualify for the BEA Program on
their own merit. Thus, the BEA Program is far more targeted than the
CRA.
As the GAO points out, the BEA Program is retrospective, rewarding
applicants for their past activities, and places no restrictions on the
use of awards. This is a statutory requirement of the BEA Program.
Although the GAO raises this as an issue, it makes no recommendation
that changes should be made to the BEA Program to require awardees to
use their award dollars for additional community development
activities. Nor does the GAO recognize that, in fact, many applicants
have stated that they do plan to use their awards to further their
community development efforts. The CDFI Fund's survey of FY 2002 BEA
applicants revealed that applicants planned to use 80 percent of their
total anticipated awards for additional community development
activities, including additional loans to CDFIs, continued financing to
underserved communities, investments in low-income businesses and
communities, and covering the operating costs of their community
development programs.
4. The GAO report also fails to mention other important program
benefits.
For example, the CDFI Fund's survey of FY 2002 BEA applicants asked if
the prospect of receiving a BEA award prompted them to launch any
innovative financial products, services or educational programs
designed to meet the needs of underserved households or communities. Of
the 104 applicants that responded to this question, 20 (19 %) responded
affirmatively.
The GAO notes that "community development loan funds have raised an
increased amount of capital from banks, thrifts and credit unions while
BEA program funding and bank participation in the program have
declined," (pg. 16). We address the issue of program funding below. The
implication of this paragraph, albeit GAO draws no conclusion in its
report, is that the BEA program may be unnecessary to promote lending
relationships between CDFIs and banks due to the size of the awards.
It would have been useful had GAO studied the underlying data to
determine whether the CDFIs that the CDFI Data Project indicates have
(1) increased their borrowings from banks which had been previous
participants in the BEA program; (2) whether the BEA Program is
responsible for initiating a positive lending relationship that has
continued; (3) or whether the BEA Program caused the banks to initiate
relationships with CDFIs that were not eligible under the BEA Program
based on their positive experience with eligible CDFIs.
5.The CDFI Fund notes that the GAO report does not correctly describe
the relative funding for the program.
Specifically, the GAO correctly notes that the budgetary authority for
the BEA Program has declined in recent years - from $46 million in FY
2000 to just under $10 million in FY 2005. However, since the CDFI
Fund's annual appropriation in its entirety has declined in recent
years, the CDFI Fund has had less budgetary authority to award. Thus,
the BEA Program has not declined as a priority for the Fund, but rather
BEA Program funding as a percentage of the total amount available for
the CDFI Fund monetary programs has stayed roughly the same - and even
increased: 28 percent in FY 2000, 33 percent in FY 2002, and 30 percent
in FY 2006.
CDFI Fund response to the GAO Recommendation:
The GAO report makes one recommendation to the CDFI Fund: to revise the
guidance for reviewing program applications so that program staff are
required to (1) geo-code property addresses where appropriate, and (2)
document their efforts to verify property addresses. In response to the
GAO's recommendation, the CDFI Fund will adopt a policy requiring
applicants to report addresses for all transactions. CDFI program staff
will be provided with updated instructions to geo-code all transactions
over $250,000 (not just all transactions over $500,000 as is current
practice) - and to provide documentation of such efforts. CDFI program
staff will initiate and implement steps to analyze a statistically
significant sample of transactions less than $250,000 to provide
additional assurance that such transactions are carried out in eligible
census tracts and will document the results.
Conclusion:
While the CDFI Fund agrees that it is difficult to measure the causal
effect a BEA Program award has on bank investments, the CDFI Fund
believes that the BEA Program is meeting its goal of encouraging banks
to increase their investments in CDFIs for lending and other financial
services in distressed communities. The CDFI Fund has provided evidence
to GAO in support of this conclusion. As our comments reflect, we are
disappointed that the GAO Report did not give fair weight to this
evidence, and instead chose to focus its findings on outdated studies
and on a very limited sample of program participants. The CDFI Fund
requests that GAO remove the discussion to the now prohibited practice
of deposit swapping. Notwithstanding our disappointment in the
methodologies employed by GAO and its findings with respect to overall
program impacts, the CDFI Fund concurs with GAO's recommendation with
respect to increasing administrative oversight of program activities,
and will move to implement this recommendation in the 2007 BEA program
round.
Sincerely,
Signed by:
Arthur A. Garcia:
Director:
Community Development Financial Institutions Fund:
The following are GAO's comments on the Department of the Treasury's
letter dated July 21, 2006.
GAO Comments:
1. Our report includes a statement by Treasury officials that the BEA
program provides banks with incentives to provide financial services in
the most distressed communities--communities that banks are not
required to service in their efforts to comply with CRA. However, as
discussed in our report, measuring the purported impact of the BEA
program is difficult.
2. Census tracts that qualify for the BEA program can exceed those
specified in Treasury's letter. For example, census tracts with poverty
rates as low as 20 percent may qualify under certain circumstances.
Therefore, the BEA program may not be as targeted as Treasury claims.
3. Our report does not address this issue. However, we note that
requiring BEA award recipients to use their award proceeds for
additional community development activities would pose complexities.
For example, it would require Treasury to develop information about
current award recipients' overall community development activities and
a mechanism for monitoring recipients' use of award dollars.
4. Our report does not comment on the BEA program's funding relative to
other related programs within Treasury. We provide information on the
program's funding for descriptive purposes only and make no assertions
concerning its priority within Treasury.
[End of section]
Appendix III: GAO Contact and Staff Acknowledgments:
GAO Contact:
George A. Scott, (202) 512-7215 or scottg@gao.gov:
Staff Acknowledgments:
In addition to the contact named above, Wesley Phillips (Assistant
Director), Emilie Cassou, David Dornisch, Ronald Ito, Austin Kelly,
Elizabeth Olivarez, David Pittman, Linda Rego, and James Vitarello made
key contributions to this report.
[End of section]
(250265):
FOOTNOTES
[1] For purposes of this report, investments in CDFIs are equity
investments, equitylike loans, grants, loans, deposits or shares, and
technical assistance.
[2] See 12 C.F.R. §1806.200, which requires a BEA award applicant to
designate one or more distressed communities in which it will carry out
distressed community financing or service activities and establishes
minimum eligibility and distress requirements for such a community.
[3] Treasury has a process for certifying a CDFI, which means that the
institution meets certain CDFI eligibility requirements--including
having a primary mission of promoting community development and a
predominant business activity of providing financial products,
development services, or other similar financing to a target population
or an investment area. 12 C.F.R. § 1805.201.
[4] As of January 1, 2006, Treasury had certified 752 CDFIs. Among
these, 55 were community development banks that FDIC insures, 146 were
credit unions that the National Credit Union Share Insurance Fund
insures and therefore are ineligible for BEA awards, 505 were loan
funds, 22 were venture capital funds, and 24 were depository-holding
companies. For purposes of this report, depository-holding companies
are considered banks.
[5] Community development banks, for purposes of this report, are those
Treasury has certified as such banks. Traditional banks, for purposes
of this report, are noncommunity development banks. BEA awards to
community development banks can be as much as three times higher than
awards to traditional banks that make similar investments and loans.
[6] GAO, Community Development: CDFI Fund Can Improve Its Systems to
Measure, Monitor, and Evaluate Awardees' Performance, GAO/RCED-98-225
(Washington, D.C.: July 15, 1998); and Office of Management and Budget,
Bank Enterprise Award Assessment (Washington, D.C., 2002).
[7] Pub. L. No. 95-128, title VIII, 91 Stat. 1147 (Oct. 12, 1977)
(codified at 12 U.S.C. §§ 2901-08). CRA requires financial regulators,
for each institution they regulate, to assess the institution's record
of meeting the credit needs of all areas in the community served,
consistent with safe and sound banking operations, and to take that
record into account in evaluating the institution's applications for a
deposit facility, such as opening new branch offices. 12 U.S.C. § 2903.
[8] Treasury's inability to award all eligible activities has resulted
in some banks' reported activities, such as increased lending in
distressed communities, not receiving BEA award dollars. For fiscal
year 2006, the Senate Committee on Appropriations expressed an
expectation that the BEA program would be funded at no less than
$11,000,000. See S. Rep. No. 109-109, 129 (July 26, 2005).
[9] S. Rep. No. 109-109, 129 (July 26, 2005).
[10] For purposes of this report, CDFI beneficiaries, also known as
CDFI partners, consist of community development banks, credit unions,
loan funds, and venture capital funds. They are the recipients of a BEA
awardee's investment.
[11] According to BEA program regulations, a distressed community is
defined as a geographic area where at least 30 percent of its residents
have incomes less than the national poverty level; the unemployment
rate is at least 1.5 times greater than the national average; and (a)
the population of that area is at least 4,000 residents if any portion
of the area is located in a metropolitan area with a population of
50,000 or greater, (b) the population must be at least 1,000 residents
if no portion of the area is located within such a metropolitan area,
or (c) the area is located entirely within an Indian reservation. 12
C.F.R. § 1806.200; and 69 Fed. Reg. 54718, 54719 (Sept. 9, 2004).
Further, under program regulations, distressed communities with poverty
rates as low as 20 percent may qualify under certain circumstances.
[12] In 2003, Treasury changed the baseline and assessment periods from
6 months each to 12 months each.
[13] There is currently a $500,000 cap on the award any one bank may
receive in a given year.
[14] That is, the bank would be eligible for $54,000 for making
$300,000 in insured deposits in the credit unions ($300,000 x 18
percent), $135,000 for increased small-business lending ($500,000 x
weighting factor of 3.0 = $1.5 million x 9 percent = $135,000), and
$180,000 for increased affordable housing lending ($1 million x
weighting factor of 2.0 = $2 million x 9 percent = $180,000). In sum,
$54,000 + $135,000 + $180,000 = $369,000.
[15] That is, the bank would receive $18,000 for $300,000 in insured
deposits ($300,000 x 6 percent), $45,000 for $500,000 small-business
lending ($1.5 million x 3 percent = $45,000), and $60,000 for
affordable housing lending ($2 million x 3 percent = $60,000). In sum,
$18,000 + $45,000 + $60,000 = $123,000.
[16] For example, the frequency would be no more than every 5 years for
a small bank with an outstanding rating and every year for a large bank
with less than a satisfactory rating.
[17] Other tests may be applied. A community development test is
applied for certain institutions known as wholesale or limited-purpose
banks, and the small-bank performance standards are applied in
evaluating the performance of a small bank or a bank that was a small
bank during the prior calendar year. See, for example, 12 C.F.R. §
345.21(a)(2) and (3) (FDIC).
[18] GAO/RCED-98-225.
[19] Board of Governors of the Federal Reserve System, The Performance
and Profitability of CRA-Related Lending (Washington, D.C., July 17,
2000).
[20] One limitation of this report is that no small banks (those with
less than $950 million in assets) responded to the report's survey and
only 21 percent of banks with $950 million to $5 billion in assets
responded.
[21] The CDFI Data Project, Providing Capital, Building Communities,
Creating Impact, Fiscal Year 2003, 3rd ed; and Providing Capital,
Building Communities, Creating Impact, Fiscal Year 2004, 4th ed. Loan
funds are typically nonprofit organizations that provide financing and
development services to businesses, organizations, and individuals in
low-income communities. There are about 500 Treasury-certified loan
funds.
[22] Treasury reports results for this measure in its annual
Performance and Accountability Report.
[23] For a more thorough discussion of criteria for effective
performance measures, see GAO, The Results Act: An Evaluator's Guide
for Assessing Agency Performance Plans, GAO/GGD-10.1.20 (Washington,
D.C.: April 1998).
[24] BEA application materials may contain both the address of the
borrower and the address of the property financed through reported bank
lending activities.
[25] GAO/RCED-98-225; and the Board of Governors of the Federal Reserve
System, The Performance and Profitability of CRA-Related Lending
(Washington, D.C., July 17, 2000).
[26] The CDFI Data Project, Providing Capital, Building Communities,
Creating Impact, Fiscal Year 2003, 3rd ed; and Providing Capital,
Building Communities, Creating Impact, Fiscal Year 2004, 4th ed.
[27] GAO, The Results Act: An Evaluator's Guide for Assessing Agency
Performance Plans, GAO/GGD-10.1.20 (Washington, D.C.: April 1998);
Agency Performance Plans: Examples of Practices That Can Improve
Usefulness for Decisionmakers, GAO/GGD/AIMD-99-69 (Washington, D.C.:
Feb. 26, 1999); Standards for Internal Control in the Federal
Government, GAO/AIMD-00-21.3.1 (Washington, D.C.: November 1999).
GAO's Mission:
The Government Accountability Office, the investigative arm of
Congress, exists to support Congress in meeting its constitutional
responsibilities and to help improve the performance and accountability
of the federal government for the American people. GAO examines the use
of public funds; evaluates federal programs and policies; and provides
analyses, recommendations, and other assistance to help Congress make
informed oversight, policy, and funding decisions. GAO's commitment to
good government is reflected in its core values of accountability,
integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through the Internet. GAO's Web site ( www.gao.gov ) contains
abstracts and full-text files of current reports and testimony and an
expanding archive of older products. The Web site features a search
engine to help you locate documents using key words and phrases. You
can print these documents in their entirety, including charts and other
graphics.
Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as "Today's Reports," on its
Web site daily. The list contains links to the full-text document
files. To have GAO e-mail this list to you every afternoon, go to
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order
GAO Products" heading.
Order by Mail or Phone:
The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or
more copies mailed to a single address are discounted 25 percent.
Orders should be sent to:
U.S. Government Accountability Office
441 G Street NW, Room LM
Washington, D.C. 20548:
To order by Phone:
Voice: (202) 512-6000:
TDD: (202) 512-2537:
Fax: (202) 512-6061:
To Report Fraud, Waste, and Abuse in Federal Programs:
Contact:
Web site: www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov
Automated answering system: (800) 424-5454 or (202) 512-7470:
Public Affairs:
Jeff Nelligan, managing director,
NelliganJ@gao.gov
(202) 512-4800
U.S. Government Accountability Office,
441 G Street NW, Room 7149
Washington, D.C. 20548: